RU2150167C1 - Method for detecting ground fault in insulated- neutral three-phase supply mains - Google Patents

Abstract

FIELD: power engineering. SUBSTANCE: method is used to detect ground fault in underground mains rated up to 1140 V. Phase failure is detected by coincidence of two results. Both results are updated once within complete phase of zero-sequence voltage variation but at relative shift of 180 deg. In total, frequency doubling of result obtained in ensured. First and second results are determined by detecting correspondence of noninverted and inverted zero-sequence voltages to first and second reference phase intervals of initial phase. First reference phase interval is set to be not over 120 deg. , second one, to equal first interval in magnitude and shifted from the latter through 180 deg. Results are considered true provided two zero transmissions of zero-sequence voltage coincide during their repetition interval. Results are determined by signal indicating hazardous ground leakage shaped at time delay in response to increase in zero-sequence voltage. In some cases, delay time is set as decreasing function of zero- sequence voltage. EFFECT: improved reliability of ground fault detection dispensing with its validity check; improved speed of serious fault detection at adequate noise immunity. 2 cl, 4 dwg

Description

 The invention relates to the field of electrical engineering, in particular, electrical safety and is intended to provide protection against electric shock and other dangerous consequences of earth leakage by identifying and subsequent protective shunting of the damaged phase of the network to earth. The invention is designed primarily for use in underground three-phase electrical networks with an isolated neutral voltage of up to 1140 V.

 In the known methods for determining the damaged phase to earth, both phase and amplitude parameters of the vibrational signals are used. The use of phase parameters provides higher accuracy of signal processing, since useful information is extracted with respect to their zero transitions, which prevents the occurrence of recognition errors of small changes in amplitude parameters for a limited time from electrical safety conditions.

A known method for determining the damaged phase of the network, implemented in [1], in which the voltage of the zero sequence is measured (using sensors), one of the line voltage of the network and the phase difference between them is the initial phase of the voltage of the zero sequence, the correspondence is controlled (using the threshold signal splitter ) values of the initial phase of the residual voltage phase reference intervals the size of 90 ^o, specify for each network phase and comprising an initial phase value corresponding pit conductive phase voltage, and determine the faulty phase network - the network is fixed as the phase damaged by detecting corresponding.

The method has a low sensitivity to leaks due to the limitation of the reference phase gaps of 90 ^o . The real value of the area of change of the initial phase of the zero-sequence voltage during single-phase dangerous leaks in the network can exceed 90 ^o under the influence of the natural asymmetry of the insulation resistances in the phases of the network. This is especially evident in high-resistance leaks - leaks with a resistance close to the asymmetric insulation resistance. As a result, the values of the specified initial phase go beyond the reference phase gap, the damaged phase of the network is not fixed. To increase the sensitivity to leaks, the expansion of the reference phase gaps is required.

In the closest to the proposed method for determining the damaged phase to earth in a three-phase network with an isolated neutral, implemented in [2], the voltage of the zero sequence is measured (the zero sequence voltage measuring unit connected between the artificial network zero point and the ground) is formed with a delay (by delay element) a signal about the occurrence of a dangerous leak to the ground to increase the voltage of the zero sequence (detected by the amplitude selector), and also for each phase of the network set the reference phase n interval of no more than 120 ^o (using the help of pulse-former shapers, called informational in the original, which convert the supply phase voltages to pulses, the duration of which sets the reference phase gaps), which includes the value of the initial phase of the supply phase voltage, control the correspondence of the reference phase gap to the initial phase of the zero sequence voltage (using a D-trigger that stores the value of the reference pulse along the front of the clock pulse, formed from apryazheniya zero sequence) and determine the damaged network phase - fixed network phase as damage, the detection of said compliance with a signal on the occurrence of hazardous leakage to ground (logical AND gate, which reveals coincidence of the detected matching time and signal the appearance of dangerous earth leakage).

The method allows you to assign the reference phase intervals with the maximum possible value of 120 ^o , when exceeded which violates the uniqueness of determining the damaged phase in a three-phase network. This provides the required sensitivity to hazardous earth leakage under electrical safety conditions against the background of possible natural asymmetry of insulation resistances in the network phases. However, this reduces the reliability of the method due to the deterioration of stability against erroneous determination of the damaged phase with a high resistance dangerous earth leakage. The deterioration of this stability is due to the compulsorily small amount of the delay in generating a signal about the occurrence of a dangerous leak to the ground. So, the delay in the method is the main means of ensuring stability against erroneous determination of the damaged phase and stability against false alarms - determining the phase of the network as damaged under the influence of impulse noise in the absence of a dangerous leak. The lower limit of the delay value is determined by the condition of stability against false positives. The upper limit is determined by the electrical safety conditions and is calculated on the basis of the permissible time of existence in the network (before the protective bypass) of the current through the person (with minimal resistance) minus the response time of the actuators. For networks with a voltage of 1140 V, the delay value can be allowed up to two periods of oscillation of the industrial frequency. With a higher mains voltage, the delay, of course, should be less. The appearance of high-resistance single-phase hazardous leaks may be accompanied by a transient with a duration exceeding the specified delay value. Such a transient process, in conjunction with other non-industrial frequency noise, can “lead away” the result of counting the initial phase of the zero-sequence voltage beyond the boundaries of the reference phase gap related to the damaged phase of the network. As a result, the neighboring phase of the network is erroneously defined as damaged. The elimination of the considered disadvantage is possible by checking the result of determining the damaged phase for reliability. However, a simple repetition of the operations of the method for the purpose of such verification is unacceptable due to the additional time costs that are unacceptable in determining the damaged phase of the network with the low resistance, most dangerous leakage to earth.

 The basis of the invention is the task of increasing the reliability of the method for determining the damaged phase to earth in a three-phase network with an isolated neutral by checking the reliability of the result of determining the damaged phase without additional time costs by increasing the frequency of obtaining the result from one to two times for a complete voltage fluctuation of the zero sequence, as well as the implementation, before determining the damaged phase to earth, of a delay dependent on leakage resistance. The use of the invention will provide stability against erroneous determination of the damaged phase without loss of speed with the most dangerous earth leakages, and will also increase speed while maintaining stability against false alarms.

To do this, in the known method for determining the damaged phase to earth in a three-phase network with an isolated neutral, in which the zero-sequence voltage is measured, a signal is generated with a delay about the occurrence of a dangerous earth leak to increase the zero-sequence voltage, and a reference phase gap is set for each phase of the network value of not more than 120 ^o, comprising a supply phase voltage value of the initial phase, the reference phase matching control interval value of the initial phase voltage Nia zero sequence determined damaged network phase - fixed network phase both damaged, with a signal of the appearance of dangerous ground fault, additionally for each phase of the network define a second reference phase interval shifted by 180 ^o with respect to the first and equal to it in magnitude, controls the second correspondence to the second reference phase gap of the value of the initial phase of the inverse voltage of the zero sequence, and fix the phase of the network as damaged, upon detection of the first and second mentioned with correspondence for one interval of succession of two zero voltage transitions of the zero sequence.

 These distinctive essential features are sufficient in all cases to which the scope of legal protection of the invention applies. The following are the distinguishing features characterizing the invention in a particular case.

 An option is proposed for setting the value of the delay in generating a signal about the occurrence of a dangerous leak to the earth with a decreasing function of the zero sequence voltage value.

Distinctive features of the new method, sufficient in all cases to which the scope of legal protection applies, are for each phase of the network:
- the task of the second reference phase gap, offset by 180 ^o relative to the first and equal to it in magnitude;
- control of the second correspondence to the second reference phase gap of the value of the initial phase of the inverse voltage of the zero sequence;
- an additional condition for determining the damaged phase of the network - by detecting the first and second mentioned correspondences in one interval of succession of two zero zero-voltage voltage transitions.

 The detection of only one match, which occurs first in time, is a condition for obtaining the preliminary result necessary to determine the damaged phase of the network. The discovery for the same phase of the network of a different correspondence, following the first one - during one interval of successive two zero zero voltage transitions of zero sequence, confirms the reliability of the preliminary result and is a condition for obtaining the final result - determining the damaged phase. Each of these conditions is detected once for a complete voltage fluctuation of the zero sequence with a relative time offset of half the full fluctuation. In aggregate, information is updated twice for the full voltage fluctuation of the zero sequence, which allows you to determine the damaged phase of the network with verification for reliability without additional time. Due to the validation check, stability against erroneous determination of the damaged phase is achieved.

 In the particular case, setting the delay value of the signal about the occurrence of a dangerous leak to the ground by a decreasing function of the zero sequence voltage value further increases the reliability of the method. The magnitude of the zero sequence voltage indirectly indicates the magnitude of the leakage resistance. A smaller value of leakage resistance corresponds to a larger value of the zero sequence voltage. Therefore, the set delay value corresponds to an increasing function of the leakage resistance value. This allows, if there is a reliability check, to determine the damaged phase of the network with a low resistance dangerous earth leak with a minimum delay, and with a high resistance dangerous leak with a longer delay, sufficient to maintain stability against false alarms with impulse noise.

 In FIG. 1 and FIG. 2 is a vector voltage diagram explaining the arrangement of the first and second phase reference gaps, respectively, as well as the operation of the method for a case of phase A fault in the network; FIG. 3 is an example block diagram of a device implementing the method; in FIG. 4 is a flowchart of a programmable microcontroller operating algorithm included in the flowchart of the device of FIG. 3.

векторы питающих фазных напряжений;

прямой и инверсный векторы напряжения нулевой последовательности при повреждении в сети фазы A;
φ₁ и φ₂ - начальные фазы векторов

отсчитываемые как фазовые сдвиги этих векторов относительно вектора

питающего напряжения фазы A;
N - нейтраль или искусственная нулевая точка сети;
∠A₁NX₁, ∠B₁NY₁, ∠C₁NZ₁ - углы, соответствующие первым опорным фазовым промежуткам;
∠A₂NX₂, ∠B₂NY₂, ∠C₂NZ₂ - углы, соответствующие вторым опорным фазовым промежуткам;
δ ≤ 120^o - величина опорных фазовых промежутков;
β - угол сдвига начальных значений первых опорных фазовых промежутков относительно векторов

входящих в промежутки.The vector diagrams show:
A, B, C - designation of network phases;

vectors of supply phase voltages;

direct and inverse voltage vectors of zero sequence in case of damage in the phase A network;
φ ₁ and φ ₂ are the initial phases of the vectors

measured as phase shifts of these vectors relative to the vector

phase A supply voltage;
N - neutral or artificial network zero point;
∠A ₁ NX ₁ , ∠B ₁ NY ₁ , ∠C ₁ NZ ₁ - angles corresponding to the first reference phase gaps;
∠A ₂ NX ₂ , ∠B ₂ NY ₂ , ∠C ₂ NZ ₂ - angles corresponding to the second reference phase gaps;
δ ≤ 120 ^o - the value of the reference phase gaps;
β is the angle of shift of the initial values of the first reference phase gaps relative to the vectors

falling in between.

An example block diagram of the device includes a phase 1 supply voltage sensor 1, a zero sequence voltage sensor 2, a reference signal zero comparator 3, an information signal zero comparator 4, a half-wave rectifier 5, an integrating link 6, a threshold element 7, and a programmable microcontroller 8. On A block diagram of the device is also shown:
U _A is the supply voltage of phase A;
U ₀ - voltage zero sequence;
z _A and z ₀ - reference and information logic signals;
s is a logical signal about the occurrence of a dangerous leak to the ground;
Q _A , Q _B , Q _C - output logic signals of the device.

The block diagram of the programmable microcontroller operation algorithm consists of functional and logical operators 9-30. The flowchart also shows:
R is the result of a preliminary determination of the damaged phase of the network;
φ is the phase shift of the current zero voltage transition U ₀ relative to the zero transition from negative to positive voltage values U _A.

R _A , R _B , R _C - network phase identifiers.

 The essence of the method is as follows.

Each phase of the network is assigned two equal in magnitude of the reference phase gap - the first and offset from it by 180 ^o second. These phase gaps are reference for monitoring the compliance with them of the values of the initial phases φ ₁ and φ ₂ of the zero sequence voltage U ₀ and its inversion -U ₀ . The location and size of the reference phase gaps are set so that, according to the correspondence of the first reference phase gap, the values of the initial phase φ ₁ (hereinafter the first correspondence) and the second reference phase gap, the values of the initial phase φ ₂ (hereinafter the second correspondence) determine the damaged phase of the network with validation check.

показанный штрихпунктиром (конец вектора соответствует потенциалу земли). Под влиянием естественной несимметрии сопротивлений изоляции в фазах сети (среди величин, не вызывающих срабатывания аппаратуры защитного отключения сети) область изменения начальной фазы напряжения U₀ расширяется. Задание первых опорных фазовых промежутков сводится к заданию одного из них и получению двух соседних путем смещения их на 120^o и 240^o соответственно. При задании первого опорного фазового промежутка его середину совмещают со средним значением области изменения начальной фазы напряжения U₀, равным - 0,5ψ₀ (знак минус указывает на отставание напряжения U₀ относительно напряжения U_A). Это позволяет в максимальной степени охватить возможное расширение области изменения начальной фазы напряжения U₀ при однофазной утечке на землю в сети с несимметричным сопротивлением изоляции фаз. В этом случае опережающие границы (начальные значения) первых опорных фазовых промежутков отстоят от соответствующих векторов питающих фазных напряжений

на угол β = 0,5δ-0,5ψ₀,δ≥ψ₀.In FIG. The angles ∠A ₁ NX ₁ = ∠B ₁ NY ₁ = ∠C ₁ NZ ₁ = δ correspond to the first reference phase gaps ₁ . Each angle is set in such a way as to cover the area of change of the initial phase of the voltage U ₀ during leakage in the corresponding phase of the network. The value of this region with a symmetric insulation resistance of the phases of the network with an isolated neutral - ψ ₀ is known in advance and does not exceed 90 ^o with an arbitrarily large value of the insulation capacity. For phase A, this region is limited by the hodograph of the vector

shown by a dash-dotted line (the end of the vector corresponds to the potential of the earth). Under the influence of the natural asymmetry of the insulation resistances in the phases of the network (among the values that do not cause the circuit breaker to trip), the region of change in the initial phase of the voltage U ₀ expands. The task of the first reference phase gaps is reduced to the task of one of them and obtaining two neighboring ones by shifting them by 120 ^o and 240 ^o, respectively. When setting the first reference phase gap, its middle is combined with the average value of the region of variation of the initial phase of the voltage U ₀ equal to 0.5ψ ₀ (a minus sign indicates a voltage lag U ₀ relative to the voltage U _A ). This allows to maximally cover the possible expansion of the region of variation of the initial phase of the voltage U ₀ during a single-phase earth leakage in a network with asymmetric phase insulation resistance. In this case, the leading boundaries (initial values) of the first reference phase gaps are separated from the corresponding vectors of the supply phase voltages

angle β = 0.5δ-0.5ψ ₀ , δ≥ψ ₀ .

расположен внутри угла ∠A₁NX₁, то есть значение начальной фазы φ₁ соответствует первому опорному промежутку фазы A. Обнаружение этого соответствия является предварительным результатом, необходимым для определения поврежденной фазы сети. В целях достижения устойчивости против ошибочного определения поврежденной фазы сети осуществляется проверка достоверности предварительного результата, то есть его подтверждение. Для этого в способе используют вторые опорные фазовые промежутки, смещенные относительно первых на 180^o и равные им по величине. На фиг. 2 им соответствуют ∠A₂NX₂= ∠B₂NY₂= ∠C₂NZ₂= δ. При повреждении фазы A вектор -

расположен внутри угла ∠A₂NX₂, то есть значение начальной фазы φ₂ соответствует второму опорному фазовому промежутку фазы A. Обнаружение второго соответствия вслед за первым подтверждает достоверность предварительного результата и является условием получения конечного результата.When phase A is damaged, the vector

is located inside the angle ∠A ₁ NX ₁ , that is, the value of the initial phase φ ₁ corresponds to the first reference interval of phase A. The detection of this correspondence is a preliminary result necessary to determine the damaged phase of the network. In order to achieve stability against erroneous determination of the damaged phase of the network, the reliability of the preliminary result is verified, that is, its confirmation. To do this, the method uses the second reference phase gaps offset from the first 180 ^o and equal to them in magnitude. In FIG. 2, they correspond to ∠A ₂ NX ₂ = ∠B ₂ NY ₂ = ∠C ₂ NZ ₂ = δ. If phase A is damaged, the vector

is located inside the angle ∠A ₂ NX ₂ , that is, the value of the initial phase φ ₂ corresponds to the second reference phase gap of phase A. The detection of the second correspondence after the first confirms the reliability of the preliminary result and is a condition for obtaining the final result.

The fundamental possibility of verifying the reliability of determining the damaged phase without impairing performance is based on the number of possible reference points for the values of the initial phase of the voltage U ₀ . With a random voltage amplitude U _0, such reference points are its zero transitions (from negative to positive values and vice versa), that is, two points during its full oscillation. At the zero transition of one direction, the value of the initial phase φ _{1 is} controlled, and at the zero transition of the opposite direction, the value of the initial phase φ _{2 is} controlled. Therefore, at the moments of zero transitions, the mentioned correspondence is also controlled. Due to the randomness of the occurrence of a leakage to the ground, the delay in the occurrence of a zero transition in any direction relative to the moment of arrival of the signal about the occurrence of a dangerous leakage to the ground can reach the time of the complete voltage fluctuation U ₀ (period for steady-state oscillations). The specified maximum delay is typical for methods for determining the damaged phase of a network using only one match, controlled by a zero transition in one direction. It is obvious that this delay includes the moment of the zero transition of another direction, in which the second correspondence is controlled in the proposed method. Thus, the verification of the reliability of the result is achieved in the proposed method without additional time in comparison with a method that does not use such verification.

The above description of the operation of the method relates to the case of the appearance of a low-resistance, most dangerous leak in the network, accompanied by a short-term transient process of establishing the voltage U ₀ , which contributes to the rapid determination of the damaged phase of the network.

In the case of the appearance of a high-resistance dangerous leakage in the network, accompanied by a long transient process of establishing the voltage U ₀ , the time to determine the damaged phase of the network will be determined by the transient decay time until a reliable result occurs. Electrical safety conditions are also observed due to the lower value of the leakage current.

With a long transient, the following sequence of events is possible. After the detection of one match, the second match during the succession of two zero voltage transitions U ₀ may not occur in two cases: due to missing the second match - when setting the reference phase gap is less than 120 ^o , or due to the detection of the second match in adjacent phase of the network. The last match, which does not coincide with the previous one, is taken as a preliminary result of determining the damaged phase of the network with the expectation of finding the next, confirming, match in it to obtain the final reliable result.

Determination of the damaged phase of the network is carried out in the presence of a signal about the occurrence of a dangerous leak to the ground. Such a signal is formed with a delay in increasing the voltage U ₀ . The delay serves to exclude the influence of impulse noise. Obviously, with a fixed delay, this is best matched by its maximum value, acceptable under the conditions of providing protection against low-impedance hazardous earth leakage. At the same time, checking the validity of determining the damaged phase can reduce the delay and, thereby, increase the effectiveness of protection. However, such a decrease in the delay can lead to false alarms in networks with asymmetric insulation resistance - to determine the damaged phase of the network in the absence of a dangerous leak. As a result of the subsequent protective shutdown of the network (simultaneously with protective shunting), unjustified downtime of technological electrical equipment occurs. The reason for such false responses is the action of impulse noise with a long decaying transient during switching in the connection network with a braked electric motor. An increase in the speed of determining the damaged phase of the network with the most dangerous, low resistance leakage to earth without loss of stability against false alarms can be achieved by setting the delay value as a decreasing function of the voltage value U ₀ . A well-known pattern is used to set the delay value — the proportionality of the duration of the transient interference to the leakage resistance. The decreasing function is set so that large values of voltage U ₀ with a low resistance hazardous earth leak correspond to a minimum delay sufficient to exclude the influence of short-term interference, and lower values of voltage U ₀ with a high resistance dangerous leakage correspond to a longer delay sufficient to exclude the influence of long interference. The magnitude of the greater delay is several times less than the maximum permissible value from the safety conditions for high resistance dangerous earth leakage. This margin of latency simplifies its implementation. A decreasing function can be specified as a strictly decreasing or step function depending on the implementation - analog or digital.

For the uniqueness of the results obtained, the value of the reference phase gaps should be no more than 120 ^o in order to avoid overlapping among the first and separately second phase gaps. To achieve maximum sensitivity to leaks, the value δ of the reference phase gaps must be assigned a maximum equal to 120 ^o . An increase in the reference phase gap also improves the dynamics of determining the damaged phase, since a reliable result is faster generated under the influence of interference, in particular, at the stage of transient processes (caused by damage) that shift zero voltage transitions U ₀ in time. The limitation of sensitivity to leaks at the maximum value of the reference phase gaps can be achieved by regulating the sensitivity of the formation of a signal about the occurrence of a dangerous leak to the ground.

The reference phase gaps and the values of the initial phases can be counted relative to any combination of the supply voltage of the network. For each phase of the network, the reference can be performed relative to the supply phase voltage or any other signal at the frequency of the network, shifted from it by a fixed angle. The value of this angle can be any, because in the samples it is numerically added both to the values of the boundaries of the reference phase gaps and to the values of the initial phase of the voltages U ₀ and -U ₀ . This does not affect the control of correspondence between the reference phase gaps of the values of the initial phases φ ₁ and φ ₂ , since as a result of comparing the latter with the boundaries of the reference phase gaps, equal additives are mutually destroyed.

When counting the reference phase gaps and the initial phases of the voltages U ₀ and -U ₀ relative to the supply phase voltage individually for each phase of the network, the result of determining the damaged phase is generated independently for each phase of the network. The advantage of individual readings is their independence from the order of the phase sequence of the network.

The phase sequence of the network, as a rule, is known in advance, and is easily checked if necessary. In this case, the reference phase gaps and the initial phases of the voltages U ₀ and -U ₀ can be counted for all phases of the network relative to any of the supply voltages or the signal generated from them. The value of its initial phase can be any, because, as already noted, it is mutually compensated in the control of the considered correspondences. The advantage of this reading is the minimum number of voltages used.

Consider an example of a device for implementing the method in which the reference phase gaps and the initial phases of voltages U ₀ and -U ₀ are counted relative to one signal — the supply voltage of phase A. In the device (Fig. 3), a signal proportional to the supply phase voltage U _A is measured sensor 1 (e.g., in the form of a three-phase transformer secondary winding, the primary windings of which are connected to the phases of the network and the network forming an artificial zero point) and the voltage U ₀ - 2 tow (made, for example, in the form of a single-phase separation of the transformer is connected between an artificial network of zero-point and the earth). From the voltage U _{A, the} null comparator 3 generates a logical reference signal z _A. From voltage U _{0, the} null-comparator 4 generates a logical information signal z ₀ , and two-half-wave rectifier 5, an integrating link 6 and threshold element 7 are connected in series by a signal s about the occurrence of a dangerous earth leak. From the input signals z _A , z ₀ and s, the programmable microcontroller 8 generates output logic signals Q _A , Q _B and Q _C. A high logical level of one of them indicates a phase A, B, or C network failure, respectively.

The signal s about the occurrence of a dangerous earth leak is generated with a delay determined by the time the element 7 threshold is reached by the output signal of the integrating link 6 when the voltage U ₀ is increased - the DC component at the output of the rectifier 5. This time is a strictly decreasing function of the voltage value U ₀ . The integrating element 6 in the simplest embodiment can be made in the form of a low-pass filter of the first order. The sensitivity to leaks is regulated by the threshold of the element 7, and the delay scale is regulated by the time constant of the integrating link 6.

в направлении хода часовой стрелки (фиг. 1, 2). Элементы 19 и 20 контролируют соответствие начальных фаз φ₁ и φ₂ соответственно первым и вторым опорным фазовым промежуткам, путем сопоставления их с границами опорных фазовых промежутков. Попадание векторов

между опорными фазовыми промежутками (при δ < 120^° ) приводит к обнулению предварительного результата R элементом 21 и переходу к элементу 10. Попадание же указанных векторов в опорные фазовые промежутки включает в работу один из логических элементов 22-24. При повреждении фазы A логический элемент 22 сравнивает результат R с идентификатором R_A фазы A. При несовпадении их данный идентификатор запоминается элементом 25 как предварительный результат определения поврежденной фазы и выполнение программы возвращается к элементу 10. С наступлением очередного нулевого перехода напряжения U₀ логический элемент 22 выявляет совпадение текущего результата с предшествующим результатом R, обращается к элементу 28, который назначает выходному сигналу Q_A значение "1". При повреждении в сети фазы B или C в работу вступают элементы 23, 26, 29 или 24, 27, 30. В качестве идентификаторов фаз сети используются двоичные коды, например: R_A = 001, R_B = 010, R_C = 100.The algorithm of the microcontroller 8 is built on the principle of software polling of the input signals. The algorithm is based on the control of the initial phases φ ₁ and φ ₂ by recalculation into degrees of time intervals between a positive zero voltage transition U _A (transition 0/1 of the signal z _A ) and, respectively, positive and negative zero voltage transitions U ₀ (transitions 0/1 and 1 / 0 signal z ₀ ). Recalculation is carried out in relation to the duration of the voltage period U _A. An internal timer is used to measure time intervals and period. In the first step of operation of the microcontroller 8, the timer is assigned the initial value corresponding to the nominal value of the period (Fig. 4, element 9). At the same time, the values of the result R of the preliminary determination of the damaged phase and the output signals Q _A , Q _B and Q _C are cleared. Elements 10 and 11 of the flowchart are used to detect the transition 0/1 of the signal z _A , by which the period T value measured by the timer is stored and the timer is restarted to a new dimension (element 12). At the first access to the timer, the nominal value of the period is read as such a value. Between timer restarts, a signal s is polled for a dangerous earth leak. If there is no leak in the network (s = 0), the operation algorithm returns to element 10. If a leak occurs in the network (s = 1), elements 15 and 16 control the appearance of zero voltage transitions U ₀ , and element 17 calculates the value in degrees of phase shift φ current transition 0/1 or 1/0 of signal z ₀ relative to transition 0/1 of signal z _A. Next, element 18 sets the direction of transition of the signal z ₀ and identifies the calculated phase shift φ with the value of the initial phase φ ₁ or φ ₂ . The reference phase gaps and the initial phases φ ₁ and φ _{2 / are} counted relative to the vector

in the direction of the clockwise movement (Fig. 1, 2). Elements 19 and 20 control the correspondence of the initial phases φ ₁ and φ _2, respectively, to the first and second reference phase gaps, by comparing them with the boundaries of the reference phase gaps. Hit vectors

between the reference phase gaps (at δ <120 ^° ) leads to the zeroing of the preliminary result R by element 21 and the transition to element 10. The falling of these vectors into the reference phase gaps includes one of the logical elements 22-24. If phase A is damaged, logic element 22 compares the result of R with identifier R _{A of} phase A. If they do not match, this identifier is stored by element 25 as a preliminary result of determining the damaged phase and program execution returns to element 10. With the next zero voltage transition U _0, logic element 22 detects the coincidence of the current result with the previous result R, refers to element 28, which assigns the output signal Q _{A to the} value "1". If a phase B or C network is damaged, the elements 23, 26, 29 or 24, 27, 30 come into operation. Binary codes are used as identifiers of the network phases, for example: R _A = 001, R _B = 010, R _C = 100.

Sources
1. A.S. USSR N 1379857, H 02 H 3/16, publ. in 1988, BI N 9.

 2. A.S. USSR N 943959, H 02 H 3/16, publ. in 1982, BI N 26.

Claims (2)

1. A method for determining the damaged phase to earth in a three-phase network with an isolated neutral, in which the voltage of the zero sequence is measured, a signal is generated with a delay about a dangerous earth leak to increase the voltage of the zero sequence, and for each phase of the network, the reference phase gap is set to not more than 120 ^o , which includes the value of the initial phase of the supply phase voltage, control the compliance of the reference phase gap values of the initial phase of the voltage zero voltage In particular, they determine the damaged phase of the network - fix the phase of the network as damaged, if there is a signal about the occurrence of a dangerous leak to the ground, characterized in that for each phase of the network a second reference phase gap is set, offset by 180 ^o relative to the first and equal in magnitude to it, control the second correspondence to the second reference phase gap of the value of the initial phase of the inverse voltage of the zero sequence, and fix the phase of the network as damaged upon detection of the first and second mentioned matches for one interval of succession of two zero voltage transitions of the zero sequence.  2. The method according to claim 1, characterized in that the delay value of the signal generation of the occurrence of a dangerous leak to the ground is set by a decreasing function of the voltage value of the zero sequence.

Images